40 ■ CITIES AND CLIMATE CHANGE
were incorporated. For Paris, the city’s Bilan Carbone (Mairie de Paris 2009)
reported that national data from the food industry were used to determine aver-
age per capita consumption; however, methods used for estimating cement and
steel consumption were not fully detailed nor were data sources for the requisite
emission factors. In Paris, detailed analysis of resident airline travel and freight
transport was conducted; both incoming and outgoing trips were counted and
fully allocated to Paris. In contrast, Ramaswami and others (2008) applied a
50 percent allocation to destination and origin locations for allocating both air-
line and surface transport in Denver; such an origin-destination allocation pro-
cedure ensures that the same trip is not double counted at both ends of a trip.
For key urban materials in U.S. cities, Ramaswami and others (2008) used
tools from industrial ecology—material fl ow analysis (MFA) and life cycle assess-
ment (LCA). MFA for Scope 3 material fl ows in cities oft en using monetary
consumption data available at the metropolitan spatial scale. Th ese fl ows are
calibrated with national material consumption data to ensure no methodologi-
cal double counting for materials occurs. Emission factors for the various mate-
rials considered for Denver—food, cement, and transport fuels (gasoline and
diesel)—were obtained from nationally calibrated LCA tools such as Carnegie
Mellon University’s Economic Input Output-LCA (Green Design Institute 2006),
the U.S. National Renewable Energy Laboratory’s Life Cycle Inventory database
(NREL 2009), and the U.S. Argonne National Laboratory’s GREET model (ANL
2009) for transportation fuels. With the exception of food—which is a complex
supply chain—embodied energy and GHG emissions from industrial production
of materials such as cement, steel, and petroleum fuels could be readily computed
for the United States. Th e IPCC provides specifi c guidance on these parameters
for international applications, particularly for nonenergy-related industrial emis-
sions from cement production and the like (IPCC 1997, 2006).
Th e review indicates that methods that avoid double counting exist and have
been applied to assess upstream impacts of key Scope 3 consumption activities
in cities. International GHG emissions data for most of these materials exist or
can be researched (for example, IPCC nonenergy emissions), with the excep-
tion of food (see also the work of Birch, Barrett, and Wiedmann 2004 in the
United Kingdom). When food production activities or cement factories occur
within city boundaries, case studies in Delhi and Kolkata (Sharma, Dasgupta,
and Mitra 2002) demonstrate that the emissions from these in-boundary activ-
ities can be allocated to avoid double counting between in-boundary and out-
of-boundary activities.
Th us, careful Scope 1–2–3 accounting of GHGs at the city scale is indeed
possible. To be consistent with other GHG accounting protocols (WRI, ICLEI,
CCAR), we propose that all cities and metropolitan regions be encouraged to
report Scope 1 and Scope 2 emissions in their baselines (as well as Scope 3